# Object Detection on Habitat Gibson Generalization 1.0 (val) Source: [Object Detection on Habitat Gibson Generalization 1.0 (val)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:23:20.452Z Evaluation of Object Detection performance on the validation split of the Habitat Gibson Generalization 1.0 dataset, reporting bounding box metrics like AP50. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Object Detection | | Dataset (including any stated split) | Habitat Gibson Generalization 1.0 (val) | | Metric | AP50 (Box) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 10 of 10 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [AP50 (Box)](): Higher is better; unit: PERCENT. ## Recorded results | Method | AP50 (Box) | Result date | Report paper | | --- | --- | --- | --- | | SEAL | 40.02 | Not recorded | 68fa6b621e209131cea2dea1 | | Active Neural SLAM + Optical Flow | 35.85 | Not recorded | 68fa6b621e209131cea2dea1 | | Semantic Curiosity + Optical Flow | 35.61 | Not recorded | 68fa6b621e209131cea2dea1 | | Frontier Exploration + Optical Flow | 35.22 | Not recorded | 68fa6b621e209131cea2dea1 | | Semantic Curiosity + Self-training | 35.04 | Not recorded | 68fa6b621e209131cea2dea1 | | Pretrained Mask-RCNN | 34.82 | Not recorded | 68fa6b621e209131cea2dea1 | | Active Neural SLAM + Self-training | 34.35 | Not recorded | 68fa6b621e209131cea2dea1 | | Random Policy + Optical Flow | 33.97 | Not recorded | 68fa6b621e209131cea2dea1 | | Frontier Exploration + Self-training | 33.78 | Not recorded | 68fa6b621e209131cea2dea1 | | Random Policy + Self-training | 33.41 | Not recorded | 68fa6b621e209131cea2dea1 | ### SEAL AP50 (Box): 40.02. Result ID: 698e8414ffc3901de09e4242. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Action Policy":"Gainful Curiosity","Perception Baseline":"3DLabelProp"}. This record does not assert independent reproduction. ### Active Neural SLAM + Optical Flow AP50 (Box): 35.85. Result ID: 698e8414ffc3901de09e423f. [Reporting paper]() Recorded evaluation configuration: {"Action Policy":"Active Neural SLAM","Perception Baseline":"Optical Flow"}. This record does not assert independent reproduction. ### Semantic Curiosity + Optical Flow AP50 (Box): 35.61. Result ID: 698e8414ffc3901de09e4241. [Reporting paper]() Recorded evaluation configuration: {"Action Policy":"Semantic Curiosity","Perception Baseline":"Optical Flow"}. This record does not assert independent reproduction. ### Frontier Exploration + Optical Flow AP50 (Box): 35.22. Result ID: 698e8414ffc3901de09e423d. [Reporting paper]() Recorded evaluation configuration: {"Action Policy":"Frontier Exploration","Perception Baseline":"Optical Flow"}. This record does not assert independent reproduction. ### Semantic Curiosity + Self-training AP50 (Box): 35.04. Result ID: 698e8414ffc3901de09e4240. [Reporting paper]() Recorded evaluation configuration: {"Action Policy":"Semantic Curiosity","Perception Baseline":"Self-training"}. This record does not assert independent reproduction. ### Pretrained Mask-RCNN AP50 (Box): 34.82. Result ID: 698e8414ffc3901de09e4239. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"State":"Pretrained"}. This record does not assert independent reproduction. ### Active Neural SLAM + Self-training AP50 (Box): 34.35. Result ID: 698e8414ffc3901de09e423e. [Reporting paper]() Recorded evaluation configuration: {"Action Policy":"Active Neural SLAM","Perception Baseline":"Self-training"}. This record does not assert independent reproduction. ### Random Policy + Optical Flow AP50 (Box): 33.97. Result ID: 698e8414ffc3901de09e423b. [Reporting paper]() Recorded evaluation configuration: {"Action Policy":"Random Policy","Perception Baseline":"Optical Flow"}. This record does not assert independent reproduction. ### Frontier Exploration + Self-training AP50 (Box): 33.78. Result ID: 698e8414ffc3901de09e423c. [Reporting paper]() Recorded evaluation configuration: {"Action Policy":"Frontier Exploration","Perception Baseline":"Self-training"}. This record does not assert independent reproduction. ### Random Policy + Self-training AP50 (Box): 33.41. Result ID: 698e8414ffc3901de09e423a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Action Policy":"Random Policy","Perception Baseline":"Self-training"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.